ON METAMORPHOSIS IN PLANTS. 117 



a widely spread view of the matter. But not only is the 

 argument fallacious, but it is entirely opposed to certain 

 experimental facts. In the first place it is illogical to 

 speak of an undifferentiated organ or rudiment of an organ. 

 One would have to admit the real existence of this abstract 

 idea of a "leaf". But it is obvious that no general con- 

 ception has a concrete existence. In dealing with abstract 

 notions, we recede from the actually existent in proportion 

 as our abstraction becomes more generalised. That which 

 the organs, which we term respectively foliage, scale, or 

 sporangium-bearing leaves, have in common must thus de- 

 pend on something else than a general conception. And in 

 this case are the rudiments (anlagen) really undifferentiated 

 or not ? 



The rudiments of a leaf and of a lateral branch are 

 commonly recognisable owing- to their position long before 

 they are otherwise distinguishable. No case is known in 

 which a protuberance, which from its position should de- 

 velop into a leaf, has become changed into a shoot, nor is 

 any example of the converse forthcoming. The instance 

 of Utricularia cited above clearly shows, however, that the 

 lines of demarcation between the two series of organs as 

 distinct categories may vanish. One is therefore compelled 

 to admit that a leaf and shoot rudiment are normally 

 different ab initio. Consequently it only becomes possible 

 to assume the indifferent nature of the rudiments in the 

 same category as, e.g., in leaves and their various forms. 

 But in this we are dealing, as has already been pointed out,, 

 with pure abstractions. Direct observation, furthermore, 

 proves that the form of an organ may change during its 

 development ; that it may ultimately become something 

 different from that which it at first promised to be. Thus 

 the leaf-blade of a scale leaf aborts, the leaf-base enlarges. 

 The organ is no longer undifferentiated ; this is already 

 proved by the abortion of the one portion of it. We are 

 entitled to speak of a functional change in this case, seeing 

 that the rudiment of a definite structure (a foliage leaf) has 

 here become functionally and structurally altered into a 

 scale leaf. This illustrates one of the most frequent cases 



